Moraxella catarrhalis is a Gram-negative diplococcus that for a long time was considered a relatively harmless commensal in the respiratory tract. At present, it is the third most frequent cause of otitis media and also a significant agent in sinusitis and lower respiratory tract infections in adults with pulmonary disease. M. catarrhalis is also one of the most common inhabitants of the pharynx of healthy children.
Two decades ago, Haemophilus influenzae and M. catarrhalis were shown to display a strong affinity for both soluble and surface-bound human IgD (1). The IgD-binding seems to be paralleled by a similar interaction with surface-bound IgD at the cellular level, a phenomenon that explains the strong mitogenic effects on human lymphocytes by H. influenzae and M. catarrhalis (2-4). An IgD-binding outer membrane protein from H. influenzae (protein D) was isolated and cloned, and shown to be an important pathogenicity factor (5). However, protein D does not bind to the majority of IgD myelomas tested, and it was suggested that encapsulated H. influenzae of serotype b expresses an additional IgD receptor (6).
Early studies demonstrated that the outer membrane proteins (OMPs) from a diverse collection of Moraxella isolates exhibit a high degree of similarity (7). Investigators have primarily focused their research efforts on a selected group of proteins. Recent studies have demonstrated that the high-molecular-weight surface antigen, termed UspA or HMW-OMP, is actually comprised of two different proteins. These proteins are named UspA1 and UspA2 (8,9,10). The apparent molecular masses of these OMPs are greater than 250 kDa as determined by SDS-PAGE analysis. Reduction with formic acid yields bands of approximately 120 to 140 kDa, suggesting that the UspA proteins form an oligomeric complex composed of several monomeric subunits (11). The predicted mass of each protein, as deduced from the cloned genes, is 88 kDa and 62 kDa for UspA1 and UspA2, respectively. It is thought that the difference in the deduced mass and the mass determined using SDS-PAGE is due to a predicted coiled coil structure (9).
In a recent patent publication, an outer membrane protein of M. catarrhalis with a molecular mass of approximately 200 kDa was isolated (12). A sequence encoding a protein of approximately 200 kDa was also provided. The protein was shown to be immunogenic, but no further biological functions were presented. In addition, a 200 kDa protein is associated with hemagglutinating M. catarrhalis (13,14).
CopB is an 80 kDa surface exposed major OMP that shows a moderate antigenic conservation. In addition, OMP CD is a 46 kDa highly conserved protein with numerous surface exposed epitopes and OMP E a 47 kDa protein detected on a variety of heterologous strains. The lactoferrin-binding (LbpA and B) and transferrin-binding (TbpA and B) proteins have molecular sizes of 99-111 and 74-105 kDa, respectively.
Certain strains of Staphylococcus aureus produce immunostimulatory exotoxins such as toxic shock syndrome toxin-1 (TSST-1), staphylococcal enterotoxin A (SEA), SEB and SEC, all of which are associated with food poisoning and toxic shock syndrome (TSS). These exotoxins have been denominated as superantigens (SAg) due to their ability to activate a high frequency of T lymphocytes. SAg bind as unprocessed proteins to HLA class II molecules on APC and oligoclonally activate T cells expressing particular TCR Vβ chains. In vivo exposure to excessive amounts of SAg results in a strong cytokine production and includes IL-2, TNF-α and IFN-γ, which are associated with a toxic shock like syndrome.
Since the discovery of the first immunoglobulin-binding bacterial protein, S. aureus protein A (SpA) in 1966, this protein has been extensively characterized. The ability of SpA to bind the Fc part of IgG is well known, but SpA also binds a fraction of 1 g-molecules of all classes due to the so called ‘alternative’ binding, which represents an interaction with the variable region of certain heavy chains. All IgG-binding capacity of S. aureus has been considered to be mediated by SpA. However, the existence of a second gene in S. aureus encoding an Ig-binding protein has also been reported. Streptococcus pyogenes and Peptostreptococcus magnus are other examples of Ig-binding bacteria. S. pyogenes produces protein H belonging to the M family of proteins, and has strong affinity for the Fc region of IgG. Proteins expressed by some strains bind IgA instead of IgG or both IgG and IgA. Protein Bac or the B-antigen is an IgA-binding protein expressed by certain strains of group B streptococci. Finally, P. magnus expresses protein L that shows high and specific affinity for Ig light chains, especially k light chains, and thereby interacts with all classes of Ig.
IgD is a unique immunoglobulin that exists in both a soluble and a surface-bound form. Both forms are encoded by the same gene and are splicing products. All mature B lymphocytes have B cell receptors (BCR) consisting of membrane-bound IgD and IgM. Soluble IgD comprises approximately 0.25% of the total amount of serum-Ig. The main function of IgD seems to be as an antigen-receptor on the B cell surface in order to optimize B cell recruitment and accelerated affinity maturation. Antigen is taken up through IgD by endocytosis followed by intracellular degradation and presentation on MHC class II for T cells, which in turn are activated and produce cytokines. Hereby, T cell help is obtained including numerous cytokines (e.g. interleukin-4) and co-stimulatory molecules such as CD28.
Despite macrophages, dendritic cells, and B cells all can present antigens to T lymphocytes, the B cells are 100-fold more efficient due to the importance of the antigen-presenting immunoglobulin on the surface. An attractive strategy in order to potentiate immunization is to directly target an antigen to the B cell receptor. It was early shown that the mouse antibody-response against bovine serum albumin (BSA) conjugated to anti-IgD monoclonal antibodies was 100-fold stronger compared to BSA administration without any antibody. In parallel, it has been demonstrated that a mouse myeloma antigen incorporated into the constant region of anti-IgD-antibodies targeted to the surface-bound IgD results in an up to 1000-fold more efficient antigen presentation on MHC class II (15).
Tolerance induction can be achieved experimentally by B cell activation through the IgD BCR without any additional T cell help. It would also be possible to treat autoimmune diseases by inducing B cell anergy and thus inhibit the production of auto-antibodies. In fact, SLE-prone mice administered dextran-conjugated anti-IgD antibodies exhibit a delayed development of autoimmunity. In yet another study it was shown that B cell activation via IgD decreases a T helper 2-induced IgE response suggesting a therapy for diminishing the IgE production in severely allergic individuals by displacing the antibody response from a Th2- to a Th1-response. By targeting antigens to the B cell receptor IgD, stimulation, tolerance, and a switch from IgE-production can be achieved. In addition, polyclonal activation has been reported. The outcome is depending on the experimental model used. With different constructs including various repeating IgD-binding segments, it is possible to tailor the response.
The T cell is a significant player in the anti-tumor response since it recognizes tumor-specific antigens. However, the important T cells display commonly depressed activity in the cancer patient due to a general immunosuppression. A triggering of T helper cells would therefore be very beneficial. Vaccination against tumors using antigen presenting cells (APC) has recently been acknowledged (17). Immunization protocols with APC pulsed ex vivo with tumor antigens (peptides) have been shown to induce effective MHC class I presentation for cytotoxic T cells. It has also been demonstrated that EBV-transformed B cells are able to present melanoma antigens for tumor-infiltrating lymphocytes (TIL). In experimental models, it has also been shown that tumor cells transfected with MHC class II and B7 surface molecules, receptors that are abundant on B cells, would be a feasible approach for tumor vaccination. Interestingly, B16 melanoma bearing mice that were injected with B cells pulsed with a tumor lysate from the corresponding cell line showed a prolonged survival due to an increase in IFN-γ producing T cells. It was also demonstrated that the induced T helper cells evoked a stronger cytotoxic response against the solid tumors. Since myeloma antigen targeted to IgD induces a T cell response, the suggested approach using IgD-binding bacterial proteins conjugated to specific tumor antigens would be feasible.
To target an antigen (e.g. peptide derived from a microbe or a specific tumor) to IgD-bearing B cells in order to trigger both humoral and cellular immune responses a IgD-binding protein or a shorter IgD-binding peptide would be a very feasible vector. Several examples of successful strategies with a similar angle of approach exist. The humoral immune response in mice against bovine serum albumin (BSA) conjugated to anti-IgD monoclonal antibodies is 100-fold stronger compared to when BSA is administered alone. A recent publication by Lunde et al. (15) describes that when a myeloma-derived peptide is integrated in the constant region of anti-IgD Fab′ fragments and injected into mice, a 1,000-fold more efficient antigen presentation is achieved against the antigen in question (15). In parallel, the Ig-binding fragment of S. aureus protein A fused with cholera toxin significantly increases both systemic and mucosal immune responses 10- to 100-fold against the cholera toxin (16). Finally, in a mouse tumor model consisting of the experimentally well defined B16 melanoma, activated B lymphocytes that are pulsed ex vivo with peptides derived from the tumor tissue can evoke a stronger anti-tumor response in vivo and consequently a prolonged survival (17).